This proposal is concerned with the detection of a perilymphatic fistula (PLY). We aim to replace existing, subjective clinical observations with objective, scientific ones having higher specificity and sensitivity for the detection of PLY. We have picked three methods based upon three separate ways a PLY can manifest itself. One method involves the visual recognition of an abnormal communication (orifice) between the inner ear and middle ear and/or the presence of any fluid. A mini endoscope will be used for transtympanic middle ear observation. This procedure is less traumatic than inspection via tympanotomy. Thus, it is less likely to create artifacts such as pooling of other body fluids which can be mistaken for a fistula leakage. The usual subjective, dichotic "yes - no" observation decision will he replaced with a graded quantitative measure that can be used to characterize interobserver reliability. The second method is a system identification fistula test. It is an inferential procedure that measures a change in a physiological response that is due to the presence of the fistula. The procedure has been developed in an animal model for perilymphatic fistula with good results and we have limited, but promising, human pilot data. The third method concentrates upon detected leakage from inner to middle ear. Beta-2 transferrin, a unique endogenous substance found in perilymph and cerebral spinal fluid, will be detected using one- or two-dimensional gel electrophoresis and immunoblotting. These three methods can be used separately or can be combined to yield potentially increased detection power. One goal will be to establish the clinical criteria for middle ear exploration by the application of receiver operator characteristic (ROC) analysis to the above-mentioned methods of detecting PLY. ROC analysis will provide a scientific basis upon which individual medical decision makers can determine the optimal use of middle ear surgery screening tests based their own assessment of costs and risks involved. A continuation, based upon the outcome of the present experiments, may include randomized clinical trials.